Linkage of Census 2021 to the Longitudinal Study: beta test

The beta test stage of the project to link Census 2021 data to the Office for National Statistics (ONS) Longitudinal Study (LS) is the final test stage prior to release of the new LS Research Database. During this stage, a select group of researchers will run approved projects using the new data to assess whether the new database is fit for purpose.

The beta test stage of the project has two primary objectives, which are to: 

• test the 2021 data for accuracy, comparability, and consistency through active analytical use 

• use the feedback from test participants to inform the decision whether to release the database to the wider research community 

The benefits for researchers involved in the beta test include:  

• early access to a unique and rich data source; results can be shared beyond the accredited project team members, CeLSIUS and the ONS only once all beta test work is completed 

• the chance to disseminate research as part of the public launch of the new database 

• dedicated user support from experienced and knowledgeable teams at the ONS and the Centre for Longitudinal Study Information and User Support (CeLSIUS) 

• opportunities for knowledge exchange and impact by contributing to ONS census work

We welcome expressions of interest to participate in the beta test from all interested parties. You will need to complete and return an expression of interest form to LongitudinalStudy@ons.gov.uk by the end of January 2023. You can also contact this email address if you need any further information.

Your expression of interest should address the beta test selection criteria which are set out in Section 3: Beta test selection criteria. 

Important dates for the beta test

December 2022

Open for expressions of interest. The form will be available from this webpage or by emailing LongitudinalStudy@ons.gov.uk.

End of January 2023

Closing date for beta test expressions of interest.

End of February 2023

Selection panel meeting to select the beta test projects to take forward.

March 2023

Applicants notified of the outcome of the selection panel meeting.

End June 2024

Full application forms need to be completed by successful applicants and sent to their nominated support team. Three forms are required: the full application form, beta test LS supplementary form and ethics form.

April-September 2024

Full application goes through the standard Longitudinal Study (LS) review panels to get officially approved as a project. Once approvals are in place the projects will be set up to include all data currently available.

July 2024- may be subject to change

Updated life events data is planned to be made available to beta test projects ahead of census data being added. This includes: Birth to sample mothers, Re-entries and Widowerhoods being updated to include 2018 and 2019 data. This is as well as Births, Deaths, Immigrants and Embarkations being updated to include 2018, 2019 and 2020 data.

December 2024- may be subject to change

Live running of beta test. We now aim to start the beta test phase in December 2024. This date is dependent on data processing running smoothly and no major problems being picked up by alpha testing. We will continue to review progress and keep users informed of any changes to this date. The live running of the beta test is currently planned to last between 4 to 6 months.

Note:

The delay to beta testing is due to unexpected problems with processing life events data for 2020 and 2021 as well as resourcing issues. We have introduced some new methods for these years to help overcome the barriers.

Not all life events data will be available for beta testing and the initial launch of the updated LS Research database. The following table outlines the years available for different life events data, at the time of beta testing:

The selection panel met to review the applications received and to select the best combination of projects to satisfy a number of pre‐agreed criteria. Membership of the panel included:

• representatives from a number of important ONS business areas
• a representative from the CeLSIUS, the Economic and Social Research Council (ESRC)‐funded LS user support team based at University College London (UCL)
• a representative from each of the LS teams in Scotland and Northern Ireland
• independent academics from the community.

The criteria to be used in this selection process are summarised in this section. Applicants will be asked to identify which of the specific criteria their proposals satisfy in the beta test expression of interest form. 

Essential criteria to be met by each individual project 

1. All projects should include a longitudinal element, relating Census 2021 data to other Longitudinal Study (LS) data.  

2. All project teams should include at least one member with previous experience of using the LS, preferably in partnership with a researcher or researchers with little or no previous LS experience.  

3. All projects should quantify the effect of item imputation on at least one variable; this should be a variable derived from one of the new questions in Census 2021 if any are included in the project.  

4. All projects should produce outputs for England and Wales separately as well as combined, unless there is a clear methodological reason why this is not appropriate or possible.  

5. All projects and the researchers involved must obtain the accreditation required by the Secure Research Service. Accreditation applications should not be submitted until it is known which projects have been selected for the beta test. More information is available on our Apply for an accredited research project webpage.

Essential criteria to be met by the portfolio of selected projects 

A number of specific variables and topics must be tested across the full portfolio of selected projects. Applications involving other topics are invited, but priority will be given to supporting those that encompass one or more of the following topics, to ensure all essential analyses are completed.  

1. New questions at Census 2021 asking about sexual orientation, gender identity, and whether respondents had served in the UK armed forces. 

2. Migration and mobility within the UK, including address one year ago, second address, workplace address and NHS registration location, research related to the coronavirus (COVID-19) pandemic may be relevant here.

3. International migration, including address one year ago, month and year of arrival, intended length of stay and passport(s) held, research related to the UK exiting the EU may be relevant here.  

4. Identity and ethnicity, including national identity, main language, English proficiency and passport(s) held (all asked for the second time so the first opportunity to look at individual transitions), plus religion and ethnicity.  

5. Household composition and relationships within the household, a focus on household structures during the coronavirus pandemic may be relevant here. 

6. Occupation coding and National Statistics Socio-economic classification (NS-SEC).

7. General health and limiting long-term illness, including transitions between censuses.  

8. Caring, including transitions in caring responsibilities between censuses and analysis of the impact of the change to the response scale, research related to the coronavirus (COVID-19) pandemic may be relevant here.  

9. Mortality, including an assessment of the utility of the LS for research into increases in mortality as a result of the coronavirus pandemic.  

Additional criteria that would be welcomed on project applications

1. Projects should address themes of interest to government departments such as: the impact of the coronavirus (COVID-19) pandemic, the effects of the UK leaving the EU, and topics such as social mobility, ageing, caring, and migration.  

2. Project teams should involve collaboration between academic groups and government departments or other relevant external agencies where possible.

3. Intention to replicate the same or very similar research using one or both of our sister studies in Scotland and Northern Ireland when that data become available.